Elasticsearch highlight: how to get entire text of the field in Java client

Question

I am new to Elasticsearch. I am hoping to get highlighted field in Java client. If I run the following query in Windows prompt:

{
    "query": {
        "filtered" : {
            "query" : {
                "term" : {
                    "title" : "western"
                }
            },
            "filter" : {
                "term" : { "year" : 1961 }
            }
        }
    },
    "highlight" : {
        fields" : {
            "title" : {}
            }
        }
}

I get nice highlighted text as follows:

{
      "_index" : "book",
      "_type" : "history",
      "_id" : "1",
      "_score" : 0.095891505,
      "_source":{ "title": "All Quiet on the Western great Front", "year": 1961}
      "highlight" : {
        "title" : [ "All Quiet on the <em>Western</em> great Front dead" ]
      }
}

The highlight

  "highlight" : {
    "title" : [ "All Quiet on the <em>Western</em> great Front dead" ]
  }

can be easily converted into a Java Map object, and the "title" property contains the entire text of the matched field, which is really what I want.

However, in Java client, I get highlighted fragments, which puts different segments of highlighted text of the same field into an array of text.

Thanks and regards.

Answer 1

In the Java API the default number of fragments that are returned is 5. So if you only want one fragment to be returned you need to set that.

client.prepareSearch("book")
 .setTypes("history")
 .addHighlightedField("title")
 .setQuery(query)
 .setHighlighterFragmentSize(2000)
 .setHighlighterNumOfFragments(1);

Answer 2

You may also set the number of fragments to 0 which will display the entire field with highlighting tags. This will also ignore fragment_size.

.setHighlighterNumOfFragments(0)

Answer 3

Here is what I found and I am not sure whether it is the right or best solution. In Java client, use setHighlighterFragmentSize method:

SearchResponse sr = client.prepareSearch("book")
                .setTypes("history")
                .addHighlightedField("title")
                .setQuery(query)
                .setHighlighterFragmentSize(2000) //set it larger than the size of the field so that the only one fragment is returned and it contains the entire text of the field.

I really want to hear what experts out there say and choose their reply as the answer.

Regards.